Pectin, 40-50 g/day for two weeks administered to nine normolipidemic and hyperlipidemic patients, had no effect on serum triglycerides but did cause a significant decrease in the serum total and unesterified cholesterol of hypercholesterolemic subjects in particular. This was associated with increased excretion of fecal bile acids and total steroids and increased concentration of plasma methyl sterols. Thus, the serum cholesterol reduction by pectin appears to be caused by increased cholesterol elimination into stools as bile acids which is then balanced by enhanced cholesterol synthesis. The composition of biliary bile acids and lipids was not changed and secondary bile acids and sterols decreased inconsistently in feces. The measurement of fecal dry weight suggested that the bulk of the pectin was degraded by bacteria during passage through the intestine. Consequently fecal mass and dry weight were not consistently increased, suggesting that pectin may not be an ideal fibre for increasing fecal bulk in functional colonic disorders.